Estimation of exposure to toxic releases using spatial interaction modeling

Table 4 OLS Regression results

	VIF	Coeff.	Std. Error	t value	Pr(>\|t\|)		Coeff.	Std. Error	t value	Pr(>\|t\|)
Intercept		33.53	5.823	5.758	9.39E-09	***	35.96	1.393	25.808	< 2E -16	***
% no high sch.	6.165	-0.007	0.0416	-0.171	0.8646
% in poverty	4.306	-0.295	0.0517	-5.706	1.27E-08	***	-0.308	0.038	-8.054	1.13E-15	***
% unemployed	2.082	1.386	0.0818	16.948	< 2E -16	***	1.365	0.079	17.315	< 2E -16	***
% non-white	1.867	-0.022	0.0157	-1.384	0.1665
Appalachian	1.613	-6.539	0.6447	-10.14	< 2E -16	***	-6.318	0.591	-10.684	< 2E -16	***
College educ.	3.401	-0.396	0.0494	-8.012	1.59E-15	***	-0.396	0.039	-10.089	< 2E -16	***
Smoking rate	1.708	0.526	0.052	10.116	< 2E -16	***	0.540	0.051	10.578	< 2E -16	***
South	5.119	0.822	0.7923	1.037	0.2996
Midwest	3.821	-6.51	0.7893	-8.247	2.39E-16	***	-6.894	0.484	-14.234	< 2E -16	***
West	4.197	-1.824	0.8067	-2.261	0.0238	*	-2.219	0.613	-3.621	0.0003	***
Physicians/1000	1.675	1.518	0.2203	6.891	6.71E-12	***	1.461	0.217	6.727	2.06E-11	***
% male	1.115	0.0451	0.1122	0.402	0.6877
Risk estimate	3.772	2.9E-07	1.5E-08	20.208	< 2E -16	***	2.9E-07	1.3E-08	22.525	< 2E -16	***

Results for the multivariate ordinary least squares regression of age-adjusted lung cancer mortality versus covariates and risk estimates from lung carcinogens and related compounds calculated with the buffer model. This is shown as it minimizes the AIC across all decay functions (Table 3).
Significance codes: 0.0001 '***' 0.001 '**' 0.01 '*'
n = 3057

ISSN: 1476-072X